Image Processing Apparatus and Method Thereof

1. An image processing apparatus comprising: a segmentation unit, arranged to segment an input image into a plurality of regions as sets of pixels having similar feature amounts; a calculator, arranged to calculate a size evaluation value V of an apparent size of each region segmented by the segmentation unit based on a number of pixels S and a shape of the each region, wherein the apparent size indicates a visual size of the region in the input image; and a corrector, arranged to perform chroma emphasis correction on the input image to weaken correction as a region has a larger size evaluation value V, wherein the calculator calculates the size evaluation value V using a dimension H in a vertical direction and a dimension W in a horizontal direction of the region as the shape, and a numerical formula which uses the number of pixels S, the dimension H, and the dimension W, so that the size evaluation value V grows large when a difference between the dimension H and the dimension W becomes small.

3. The apparatus according to claim 1 , wherein the calculator uses a dimension in the vertical direction of a central portion of the region as the dimension H in the vertical direction of the region, and uses a dimension in the horizontal direction of the central portion of the region as the dimension W in the horizontal direction of the region.

4. The apparatus according to claim 1 , wherein the segmentation unit executes the segmentation using information associated with colors of an image as the feature amounts.

5. The apparatus according to claim 1 , wherein the corrector individually performs the chroma emphasis correction on the regions segmented by the segmentation unit.

6. The apparatus according to claim 1 , wherein when the number of pixels of the each segmented region is smaller than a predetermined number of pixels, the segmentation unit classifies pixels of that region which do not belong to any region, and the calculator sets the size evaluation value V of a set of pixels which do not belong to any region to be zero.

7. The apparatus according to claim 1 , wherein the corrector comprises: a coefficient calculator, arranged to calculate coefficients of chroma emphasis correction of the region for respective pixels in accordance with the size evaluation value V of each region, and to store the coefficients of chroma emphasis correction in a memory; and a filter, arranged to apply two-dimensional filtering to the coefficients of chroma emphasis correction stored in the memory.

8. An image processing method comprising: using a processor to perform the steps of: segmenting an input image into a plurality of regions as sets of pixels having similar feature amounts; calculating a size evaluation value V of an apparent size of each segmented region based on a number of pixels S and a shape of the region, wherein the apparent size indicates a visual size of the region in the input image; and performing chroma emphasis correction on the input image to weaken correction as a region has a larger size evaluation value V, wherein the calculating step calculates the size evaluation value V using a dimension H in a vertical direction and a dimension W in a horizontal direction of the region as the shape, and a numerical formula which uses the number of pixels S, the dimension H, and the dimension W, so that the size evaluation value V grows large when a difference between the dimension H and the dimension W becomes small.

10. The method according to claim 8 , wherein the calculating step uses a dimension in the vertical direction of a central portion of the region as the dimension H in the vertical direction of the region, and uses a dimension in the horizontal direction of the central portion of the region as the dimension W in the horizontal direction of the region.

11. The method according to claim 8 , wherein the segmenting step executes the segmentation using information associated with colors of an image as the feature amounts.

12. The method according to claim 8 , wherein the correcting step individually performs the chroma emphasis correction on the regions segmented in the segmenting step.

13. The method according to claim 8 , wherein when the number of pixels of the each segmented region is smaller than a predetermined number of pixels, the segmenting step classifies pixels of that region which do not belong to any region, and the calculating step sets the size evaluation value V of a set of pixels which do not belong to any region to be zero.

14. The method according to claim 8 , wherein the correcting step comprises the steps of: calculating coefficients of chroma emphasis correction of the region for respective pixels in accordance with the size evaluation value V of each region, and storing the coefficients of chroma emphasis correction in a memory; and applying two-dimensional filtering to the coefficients of chroma emphasis correction stored in the memory.